Abstract: A system and method are disclosed associated with a cloud platform as a service provider. A monitoring agent, associated with a database service instance running on a client database virtual machine, periodically executes a performance throttling detection engine. The performance throttling detection engine gathers database statistics based on metrics and features of the database service using a rule-based approach. When it is determined that a pre-determined condition is met, the system may transmit the gathered database statistics to an external application. For example, the external application might comprise a database tuning service and the pre-determined condition may be associated with a decision that the database service may potentially need to tune knob configuration parameters (associated with memory knobs, background writer knobs, asynchronous knobs, etc.).

Abstract: Provided are an in-memory management system and method using user traffic. According to embodiments thereof, traffic bandwidth is monitored for an ERP user who has left an LAN coverage area, by using an OpenFlow-based SDN WAN, and loading or unloading of a table partition set, an SQL Plan cache, and an SQL view cache in an in-memory is managed depending on a result of monitoring, so that the in-memory can be efficiently managed. The user traffic for the ERP user and the user can be monitored through the local network, and loading or unloading of table partitions of the in-memory can be managed on the basis of the monitored user traffic so that the in-memory can be managed regardless of Internet access. Data transmission rate and security can be enhanced because distributed communication is performed over an LAN and a WAN for each user and each ERP user.

Abstract: Provided are an in-memory management system and method using user traffic. According to embodiments thereof, traffic bandwidth is monitored for an ERP user who has left an LAN coverage area, by using an OpenFlow-based SDN WAN, and loading or unloading of a table partition set, an SQL Plan cache, and an SQL view cache in an in-memory is managed depending on a result of monitoring, so that the in-memory can be efficiently managed. The user traffic for the ERP user and the user can be monitored through the local network, and loading or unloading of table partitions of the in-memory can be managed on the basis of the monitored user traffic so that the in-memory can be managed regardless of Internet access. Data transmission rate and security can be enhanced because distributed communication is performed over an LAN and a WAN for each user and each ERP user.

Abstract: Embodiments allow automated provisioning of a plan upgrade for databases hosted in storage environments. A database is hosted in a shared storage environment according an existing plan, based upon consumption of available system resources (e.g., processing, I/O, memory, disk). An agent periodically issues requests for information relevant to database behavior (e.g., performance metrics, query logs, and/or knob settings). The agent collects the received information (e.g., via a domain socket), performing analysis thereon to predict whether future database activity is expected remain within the existing plan. Such analysis can include but is not limited to compiling statistics, and calculating values such as entropy, information divergence, and/or adjusted settings for database knobs. Based upon this analysis, the agent communicates a recommendation including a plan update and supporting statistics.

Abstract: Various examples are directed to systems and methods for tuning a database service in a cloud platform. A tuning service may access a neural network model trained to classify workload points to one of classes. The tuning service may execute the neural network model with a first source workload point as input to return a first class as output, where the first source workload describing a source database. The tuning service may select a target workload for the first source workload point from a plurality of reference workloads. Selecting the target workload may be based at least in part on the first class returned by the neural network model. The tuning service may generate a recommended knob configuration for the source database using the target workload.

Abstract: A host delegates Just-In-Time (JIT) bytecode compilation to a serverless Web Assembly (WASM) runtime. The WASM runtime receives the bytecode, together with any additional arguments (e.g.: offsets of dependent functions, vtable metadata, virtual machine state). The host may include a parser to provide the additional arguments. In response to receiving the bytecode and arguments, the WASM runtime triggers a thread and loads appropriate WASM modules to compile the bytecode. The resulting assembly instructions are sent back to the host for execution in connection with the (frequently requested) method. Only the bytecode of frequently-accessed methods (as determined at the host) may be delegated for compilation. Delegation of bytecodes for compilation according to embodiments, may conserve a significant percentage of CPU cycles at the host, which can then be used for executing code instead.

Abstract: A computer-implemented method includes constructing a tenancy knowledge graph having a plurality of tenant nodes representing respective tenants in a multitenant computing environment, a plurality of property nodes representing respective properties of the tenants, and a plurality of edges connecting the plurality of tenant nodes and the plurality of property nodes, transforming the plurality of property nodes to corresponding property vectors, performing random walks starting from the plurality of tenant nodes of the tenancy knowledge graph, feeding sequences of nodes traversed by the random walks into a neural network to generate a plurality of tenant vectors corresponding to the plurality of tenant nodes, and clustering the plurality of tenant nodes into one or more tenant clusters based on similarity of the plurality of tenant vectors.

Abstract: A computer-implemented method includes constructing a tenancy knowledge graph having a plurality of tenant nodes representing respective tenants in a multitenant computing environment, a plurality of property nodes representing respective properties of the tenants, and a plurality of edges connecting the plurality of tenant nodes and the plurality of property nodes, transforming the plurality of property nodes to corresponding property vectors, performing random walks starting from the plurality of tenant nodes of the tenancy knowledge graph, feeding sequences of nodes traversed by the random walks into a neural network to generate a plurality of tenant vectors corresponding to the plurality of tenant nodes, and clustering the plurality of tenant nodes into one or more tenant clusters based on similarity of the plurality of tenant vectors.

Abstract: Some embodiments may be associated with a cloud-based computing environment. A WASM runtime may execute as serverless functions on an entity (VM or container) dynamically selected based on a data store location (associated with data locality and/or gravity). The WASM runtime may include one or more sandboxes each running a WASM module. A database service may access the data store, and the database service may execute on the same entity as the WASM runtime. In some embodiments, an orchestration layer selects the entity based on a default policy or user-defined custom rules in accordance with exposed attributes (CPU load, memory load, read/write mixture, etc.). According to some embodiments, the serverless functions execute in a multi-tenant fashion. Moreover, the WASM runtime process may use instruction set secure enclaves to secure an access host such that, even if a root is compromised, an attacker cannot access a sandbox memory heap.

Abstract: Technologies are described for the automated determination of materials. For example, material can be automatically identified (e.g., unique material numbers can be determined) based on sensor data and using machine learning models. In some implementations, as part of a first phase, a first set of sensor information describing the material is obtained. Using the first set of sensor information, a material class of the material is determined. As part of a second phase, a second set of sensor information describing the material is obtained. Using the second set of sensor information, the specific material is identified (e.g., a unique material identifier is determined for the material).

Abstract: Some embodiments may be provided for an enterprise associated with on-premise systems and cloud-based computing applications. A plurality of on-premise collector agents may each execute between an on-premise system and a collector system server. A plurality of cloud collector agents may each execute between a cloud-based computing application and the collector system server. A collector and optimization framework, executing on the collector system sever, may collect connectivity interfaces data from the on-premise and cloud collector agents and store the collected connectivity interfaces data in a local data store. The stored connectivity interfaces data may then be automatically analyzed to generate an optimization recommendation for an enterprise landscape. The collector and optimizer framework may then transmit information about the optimization recommendation to a user via a User Interface (“UI”) application.

Abstract: A system and method are disclosed to facilitate a database tuning as a service offered by a cloud platform as a service provider. A throttling detection engine, associated with a database service instance, may periodically determine if an automated database tuning process should be performed. When it is determined that the automated database tuning process should be performed, the throttling detection engine may transmit database performance metrics. A database tuner as a service, coupled to the throttling detection engine, may access aggregated database performance metrics of the database service instance and determine a set of tunable parameters associated with the database service instance. The database tuner as a service may then execute the automated database tuning process to recommend, using an intelligent algorithm, a new set of configurations for the set of tunable parameters to be applied to the database service instance.

Abstract: Some embodiments may be associated with a peer-to-peer platform as a service framework. A control plane processor may push a workload associated with a client request to a peer-to-peer platform as a service in accordance with resource availability. A data plane may include a plurality of node processors, and a first node processor may receive a job from the control plane and determine if: (i) the first node processor will execute the job, (ii) the first node processor will queue the job for later execution, or (iii) the first node processor will route the job to another node processor. In some embodiments, the first node processor may provide sandboxing for tenant specific execution (e.g., implemented via web assembly).

Abstract: A system and method are disclosed associated with a cloud platform as a service provider. A monitoring agent, associated with a database service instance running on a client database virtual machine, periodically executes a performance throttling detection engine. The performance throttling detection engine gathers database statistics based on metrics and features of the database service using a rule-based approach. When it is determined that a pre-determined condition is met, the system may transmit the gathered database statistics to an external application. For example, the external application might comprise a database tuning service and the pre-determined condition may be associated with a decision that the database service may potentially need to tune knob configuration parameters (associated with memory knobs, background writer knobs, asynchronous knobs, etc.).

Abstract: A host delegates Just-In-Time (JIT) bytecode compilation to a serverless Web Assembly (WASM) runtime. The WASM runtime receives the bytecode, together with any additional arguments (e.g.: offsets of dependent functions, vtable metadata, virtual machine state). The host may include a parser to provide the additional arguments. In response to receiving the bytecode and arguments, the WASM runtime triggers a thread and loads appropriate WASM modules to compile the bytecode. The resulting assembly instructions are sent back to the host for execution in connection with the (frequently requested) method. Only the bytecode of frequently-accessed methods (as determined at the host) may be delegated for compilation. Delegation of bytecodes for compilation according to embodiments, may conserve a significant percentage of CPU cycles at the host, which can then be used for executing code instead.

Abstract: Technologies are described for the automated determination of materials. For example, material can be automatically identified (e.g., unique material numbers can be determined) based on sensor data and using machine learning models. In some implementations, as part of a first phase, a first set of sensor information describing the material is obtained. Using the first set of sensor information, a material class of the material is determined. As part of a second phase, a second set of sensor information describing the material is obtained. Using the second set of sensor information, the specific material is identified (e.g., a unique material identifier is determined for the material).

Abstract: Some embodiments may be provided for an enterprise associated with on-premise systems and cloud-based computing applications. A plurality of on-premise collector agents may each execute between an on-premise system and a collector system server. A plurality of cloud collector agents may each execute between a cloud-based computing application and the collector system server. A collector and optimization framework, executing on the collector system sever, may collect connectivity interfaces data from the on-premise and cloud collector agents and store the collected connectivity interfaces data in a local data store. The stored connectivity interfaces data may then be automatically analyzed to generate an optimization recommendation for an enterprise landscape. The collector and optimizer framework may then transmit information about the optimization recommendation to a user via a User Interface (“UI”) application.

Abstract: Some embodiments may be associated with a cloud-based computing environment. A WASM runtime may execute as serverless functions on an entity (VM or container) dynamically selected based on a data store location (associated with data locality and/or gravity). The WASM runtime may include one or more sandboxes each running a WASM module. A database service may access the data store, and the database service may execute on the same entity as the WASM runtime. In some embodiments, an orchestration layer selects the entity based on a default policy or user-defined custom rules in accordance with exposed attributes (CPU load, memory load, read/write mixture, etc.). According to some embodiments, the serverless functions execute in a multi-tenant fashion. Moreover, the WASM runtime process may use instruction set secure enclaves to secure an access host such that, even if a root is compromised, an attacker cannot access a sandbox memory heap.

Abstract: Various examples are directed to systems and methods for tuning a database service in a cloud platform. A tuning service may access a neural network model trained to classify workload points to one of classes. The tuning service may execute the neural network model with a first source workload point as input to return a first class as output, where the first source workload describing a source database. The tuning service may select a target workload for the first source workload point from a plurality of reference workloads. Selecting the target workload may be based at least in part on the first class returned by the neural network model. The tuning service may generate a recommended knob configuration for the source database using the target workload.

Abstract: Various examples are directed to systems and methods for network virtualization for web applications. For example, a WAN controller may receive from a backend query processor, a first tenant user query table that describes a first set of query types for a first user. The first set of query types may comprises a first query type having a first computational weight and a second query type having a second computational weight less than the first computational weight. The WAN controller may determine that a first network path to a web application data center is less congested that a second network path to the web application data center. The WAN controller may send a first flow entry to a first network appliance on the first network path instructing the first network appliance on the first network path to send packets associated with queries of the first set of query types to a second network appliance on the first network path.